Wanting to Be Boss and Wanting to Be Subordinate: Effects on Performance Motivation

(1)

Article

Reference

Wanting to Be Boss and Wanting to Be Subordinate: Effects on Performance Motivation

SCHMID MAST, Marianne, HALL, Judith A., SCHMID, Petra C.

Abstract

Does dyad members' motivation to take on a high or low power position influence the dyad's performance motivation when assigned to hierarchical roles? Participants in 69 dyads (33 all-women, 36 all-men) indicated whether they preferred the high-power role (owner of an art gallery) or the low power role (assistant to the owner). Power roles were then randomly assigned. The dyad's interaction during task solving was videotaped. Uninvolved coders rated performance motivation as the degree of quality of the superior's and the subordinate's task contributions and effort put into the task. Performance motivation was better if the boss preferred the high power to the low power role, irrespective of the subordinate's role preference. Leadership effectiveness is thus affected by the superior's power motivation.

SCHMID MAST, Marianne, HALL, Judith A., SCHMID, Petra C. Wanting to Be Boss and Wanting to Be Subordinate: Effects on Performance Motivation. Journal of Applied Social Psychology , 2010, vol. 40, no. 2, p. 458-472

DOI : 10.1111/j.1559-1816.2009.00582.x

Available at:

http://archive-ouverte.unige.ch/unige:101510

Disclaimer: layout of this document may differ from the published version.

1 / 1

(2)

Effects on Performance Motivation¹ MarianneSchmidMast²

Department of Work and Organizational Psychology

University of Neuchâtel Neuchâtel, Switzerland

JudithA. Hall Northeastern University

PetraC. Schmid University of Neuchâtel Neuchâtel, Switzerland

Does dyad members’ motivation to take on a high or low power position influence the dyad’s performance motivation when assigned to hierarchical roles? Participants in 69 dyads (33 all-women, 36 all-men) indicated whether they preferred the high- power role (owner of an art gallery) or the low power role (assistant to the owner).

Power roles were then randomly assigned. The dyad’s interaction during task solving was videotaped. Uninvolved coders rated performance motivation as the degree of quality of the superior’s and the subordinate’s task contributions and effort put into the task. Performance motivation was better if the boss preferred the high power to the low power role, irrespective of the subordinate’s role preference. Leadership effectiveness is thus affected by the superior’s power motivation.jasp_582458..472

The specific power position a person holds in a hierarchy only to some extent determines the person’s behavior. Within a given power position, personality factors such as personality dominance or leadership style affect behavior and performance outcome (Lord & Hall, 1992; Mann, 1959). In the present study, we are interested in how the motivation to hold a high power position (i.e., wanting to be boss) or a low power position (i.e., wanting to be subordinate) affects the motivation of a superior and a subordinate to perform well in a dyadic interaction.

Superiors differ in how they fulfill their high power roles. Some superiors embrace the power that goes with the high position, whereas others seem to

1The authors thank Gwen Coutu, Kevin Elgee, Sara Eltzroth, Yota Gikas, Jessica LeDuc, Joey Pasquino, Maranda Reynolds, Holly Salach, and Sarah Witherell for their help in running participants. We also thank Mary Ellis and Jillian Andrade for their help in coding data.

2Correspondence concerning this article should be addressed to Marianne Schmid Mast, Department of Work and Organizational Psychology, University of Neuchâtel, Rue de la Maladière 23, CH-2000, Neuchâtel, Switzerland. E-mail: marianne.schmid@unine.ch

458

Journal of Applied Social Psychology,2010,40, 2, pp. 458–472.

(3)

be reluctant when it comes to accepting the power associated with the leadership role (Chan & Drasgow, 2001). These differences in motivation to lead are likely to affect performance motivation and effort. Similarly, there are subordinates who find themselves in exactly the power position they wanted, and there are subordinates who strive for a higher power position within the hierarchy. Again, it is very likely that this difference in power motivation of the subordinate affects individual or team performance effort or motivation.

Superior and Subordinate Personality Characteristics Affect Performance Outcomes

Job performance is related to personality characteristics of the performer.

For instance, Judge and Bono (2001; Bono & Judge, 2003a) found that core self-evaluation traits (i.e., self-esteem, generalized self-efficacy, locus of control, emotional stability) predicted job performance, as did conscientious- ness (Barrick & Mount, 1991). However, it must be noted that these results did not focus on the hierarchical aspect of the work relationship (i.e., leader and subordinate).

In terms of the superior’s characteristics, there has been renewed interest in the role of individual differences in explaining leadership behavior and leadership outcome (Lord & Hall, 1992; Mann, 1959). Meta-analyses have shown that a leader’s extraversion and openness are positively related to leadership effectiveness (Judge, Bono, Ilies, & Gerhardt, 2002). Regarding team performance, certain personality characteristics (e.g., warm, friendly, self-confident, able to stand pressure) of the captain of a flight crew are related to crew performance (measured as errors; Chidester, Helmreich, Gregorich, & Geis, 1991). Also, individual differences in the superior’s transformational leadership (i.e., charismatic leadership promoting a vision that inspires, motivates, and intellectually stimulates subordinates) are positively related to some aspects of subordinate task performance (Bono & Judge, 2003b).

Transformational leadership is related to better performance outcomes, such as cost performance 1 year later (Keller, 2006). Hoyt and Blascovich (2003) found that groups working under a transformational leader showed better qualitative task performance than did groups working under a transactional leader (i.e., leadership understood as an exchange of rewards for effort). And transformational leadership has been linked to positive motivational effects on followers (Bass, 1985; Burns, 1978; Shamir, House, &

Arthur, 1993). Transformational leadership style of sports coaches is related to increased intrinsic performance motivation in athletes (Charbonneau, Barling, & Kelloway, 2001). Klimoski and Hayes (1980) showed that specific

(4)

leader behavior (e.g., support, involving subordinate in determining stan- dards) is related to increased subordinate performance effort.

Moreover, there is evidence that the characteristics of the low power person in a hierarchical relationship influence job performance. McColl- Kennedy and Anderson (2002) showed a direct influence of subordinates’

frustration and optimism on their performance.

The Role of Power Motivation

We are interested in how one’s desire to occupy the high power as compared to the low power position (i.e., power motivation) affects performance in a dyadic interaction. The idea that power motivation plays an important role in leadership is not new. Chan and Drasgow (2001) definedmotivation to leadas “an individual-differences construct that affects a leader’s or leader- to-be’s decisions to assume leadership training, roles, and responsibilities and that affect his or her intensity of effort at leading and persistence as a leader”

(p. 485). In Chan and Drasgow’s study, motivation to lead was related to ratings of leadership potential based on observations in an assessment center.

However, no other leadership effectiveness or performance measure was used in their study.

We define power motivationas an individual-difference measure of the extent to which a person is inclined to fulfill a leadership role (characterized by taking on responsibility for a joint—team or dyad—performance and for the task of evaluating his or her interaction partner) or to fulfill a follower role (characterized by assisting the leader and taking his or her orders) in a specific situation. It is operationalized by the desire to occupy a high or a low position in a hierarchy. Unlike Chan and Drasgow (2001), we understand power as a continuum, with the two poles of wanting to be the leader (high power motivation or motivation to lead) and wanting to be the subordinate (low power motivation or motivation to follow), as opposed to a more or less pronounced desire for the high power position or the absence thereof.

Note that power motivation, as we define it, is influenced by both a stable personality characteristic (e.g., personality dominance) and situational factors, such as the task at hand, the characteristics of the interaction partner, or mood. As an example, a dominant person is inclined to take on the leadership role in a dyadic interaction, but the fact that he or she does not like the task at hand might make him or her less prone to want to take on the dominant role. Therefore, we expect that preference for a high or low power role (i.e., power motivation) will be related to personality dominance, although not completely identical with it.

(5)

The Present Study

Following an expectation states theory approach, legitimation or approval of existing power hierarchies by its members is an important aspect of the stability of hierarchies and their good functioning (Berger, Ridgeway, Fisek, & Norman, 1998; Ridgeway & Berger, 1986). When a power hierarchy is not legitimated, power struggles or weak leadership are to be expected, which, in turn, are likely to affect performance outcomes negatively. When superiors do not fulfill their high power roles because they have a preference for a low power role (i.e., low power motivation), they might behave as—and, therefore, be viewed as—illegitimate, resulting in weak leadership. More specifically, we hypothesize that dyads with a leader who wants to be leader (i.e., a leader with high power motivation) will invest more effort into performing well (i.e., high performance motivation) than will dyads with a leader who wants to be subordinate (i.e., a leader with low power motivation).

As for the influence of the power motivation of the subordinate, different outcomes are possible. Subordinates with a high power motivation might put up a power struggle in the dyad that might distract from the task goal and, therefore, reduce effort. Based on leader–member exchange theory (Graen, Novak, & Sommerkamp, 1982), these subordinates might be responsible for a low quality of the superior–subordinate relationship.

Research has shown that low quality of the dyadic relationship is associated with low subordinate performance and decreased organizational commit- ment, and thus low performance motivation of the subordinate (Gerstner &

Day, 1997). On the other hand, high power-motivated subordinates might want to show that they actually deserve a high power position and put a great deal of effort into the task. This study is new because it investigates the effects of both the superior’s and the subordinate’s power motivation on performance motivation.

Method Participants

Participants were 138 undergraduate students at Northeastern Univer- sity, who were majoring in different subjects. Their mean age was 18.8 years, and they were mostly Caucasian (84%; 6% Asian Americans; 4% African Americans; 4% Latino Americans; 2% others). Students received partial course credit for their participation.

(6)

Procedure

Participants signed up for the study in slots of 2 people, and we asked that people in the same dyad should not know each other. Upon their arrival in the lab, participants’ personality dominance was assessed with two question- naires. Participants were then told that they would be engaged in an interaction in which one would be the owner of an art gallery (i.e., high power role) and the other would be the assistant to the owner (i.e., low power role). We selected a task that would be interesting to students, but in which general competence would be low; namely, a discussion about works of art (i.e., paintings). In the present study, assigned power roles were not confounded by task competence differences because participants were assigned to the power roles randomly.

We only ran same-gender dyads (33 all-female, 36 all-male). To include both same- and opposite-gender dyads would have added another factor to our study; therefore, we would have needed to increase the number of participants in order to conserve the statistical power of our analysis. We opted for investigating only same-gender dyads because in opposite-gender dyads, gender often functions as a status cue, and we did not want to confound the status attribution with another potential status cue.

Participants were instructed on what the different roles entailed. The high power role included being responsible for the art gallery, giving instructions to the assistant, and evaluating the assistant; while the low power role was defined as fulfilling the requirements of the owner.³Participants were asked to indicate separately which role they preferred (i.e., role preference) after being informed what each role entailed. Roles were then randomly assigned to participants by flipping a coin. This resulted in owners who wanted to be owner (N=26), owners who wanted to be assistant (N=43), assistants who wanted to be owner (N=36), and assistants who wanted to be assistant (N=33).

Participants then interacted in their respective roles for 10 min to get accustomed to the roles. Participants were told that their art gallery partici- pates in a nationwide contest of “Selecting the Best Art Galleries,” based on a presentation by the 2 participants. This presentation consisted of a 2-min speech about a selected piece of art. Participants were given 8 min to plan and prepare a presentation together. They were instructed that during that first

3The instructions are as follows: “As an owner of an art gallery, you represent the gallery and are responsible for the gallery’s good reputation. Therefore, you tell the assistant, whom you hired, what kind of work you would like him/her to do for you and how you would like it to be done” or “As an assistant to the owner of an art gallery, you work together with your boss and try to fulfill the job requirements he/she has . . .”. At the end of the first task, the owner will be asked to evaluate the partner’s qualities as an assistant.

(7)

task, it was important to work together, but that the owner decides what contributions of the assistant he or she wants to incorporate into the presentation. Participants were also told that the owner had final responsibility about the quality of the presentation and that the owner would be asked to evaluate the assistant’s qualities as an assistant after the task.

After this first interaction, which was aimed at familiarizing participants with their respective roles, they were separated. Each participant selected from 14 different paintings the one that he or she thought should be exhibited in the art gallery. The 2 participants were brought together again, and their task was to convince the interaction partner about their own choice of painting during an 8-min discussion. This interaction was videotaped, with their knowledge. After the interaction, participants completed a question- naire measuring felt dominance during the interaction, liking of the assigned role, and justification of their initial role preference. Participants were then debriefed and thanked for their participation.⁴

Measures

Personality dominance. To assess personality dominance, we used the control scale of the FIRO-B (Schutz, 1958; sample items include “I try to be the dominant person when I am with people” and “I try to have other people do things I want done”) and created an additional 10-item scale including 6 items of the CPI dominance scale (Gough, 1975), as well as 4 additionally created items (sample items include “I think I would enjoy having authority over other people” and “I doubt whether I would make a good leader”; the latter item was reverse-scored). The FIRO-B is a Guttman scale, and the 10-item scale was reliable (Cronbach’s a =.77). The two measures were correlated (r=.54,p<.0001), which is why we averaged them after having z-transformed them to obtain a composite personality dominance measure.

Power motivation. The preference for either the high or the low power position was the operationalization of power motivation. There were 76 participants who preferred the assistant role, and 62 who preferred the owner role. To validate the power motivation measure, we tested whether (accord- ing to our definition of power motivation) power motivation was significantly related to personality dominance. The results confirm this relation (r=.46, p<.0001). Moreover, we asked participants at the end of the study why they initially chose the owner or the assistant role and coded their responses as to whether they mentioned a reason related to the (high or low) power aspect of the role (e.g., “not wanting to take on responsibility,” when preferring the

4For a more detailed description of the entire method, see Schmid Mast and Hall (2003).

(8)

low power position; “wanting to be the one who guides the interaction,”

when preferring the high power position), rather than related to other issues, such as indifference about the role or a lack of competence concerning the topic. The vast majority of participants (86.3%) mentioned reasons related to power for their initial choice, further corroborating that the role preference actually measured power motivation.

Manipulation check. Assigning the roles of owner and assistant, respec- tively, was anticipated to make participants feel more dominant in the owner role and less dominant in the assistant role. We measured felt dominance with nine items developed by the researchers (e.g., “I felt that I was the dominant one in the interaction”; a =.77). Indeed, assigned owners reported feeling more dominant than assigned assistants,t(68)=4.69,p<.0001.

To assure that the owner and assistant roles were equally attractive to participants, we used five items (developed by the researchers) to assess role liking and asked the participants how much they liked their assigned roles (e.g., “I felt comfortable in my assigned role”;a =.78). As expected, participants in the owner role and participants in the assistant role did not differ in their role liking,t(68)=1.07,p=.29. However, participants who could take on the role they wanted liked their role better than did people who had to take on the role they did not prefer: assigned assistants,t(67)=52.11,p<.05;

and assigned owners,t(67)=52.62,p<.01.

Performance motivation. Based on the videotaped 8-min interaction, two independent observers rated the quality of the superior’s and the subordinate’s task contribution and the effort each put into the task. Performance motivation is understood as the quality of the task contributions and the effort put into the task. Evaluations were performed on two items, each of which was rated on a 6-point scale ranging from 1 (inadequate) to 6 (excel- lent). The two items are “How much effort did the owner (assistant) put into the task?” and “How high was the quality of the owner’s (assistant’s) suggestions, contributions, and ideas?” Ratings for each variable were performed for each minute of the interaction and then were averaged across the 8 min. The averaging is justified by the high stability of the ratings across the 8 min (as ranged from .81 to .83). Two trained coders (trained on the first 2 min of 5 dyads for both variables; interrater reliability ranged from .74 to .96) each coded half of the targets.

Both items were highly intercorrelated (assistants, r=.94, p=.0001;

owners,r=.96,p=.0001), which is why we averaged the scores on both items for owners and assistants separately to obtain an individual measure of performance motivation (i.e., individual performance motivation). Because the performance motivation of the owner and assistant were highly correlated (r=.62,p<.0001), we combined them to obtain a measure of dyadic performance motivation (M=3.28,SD=0.88).

(9)

Owner and assistant behavior. Because performance was based on coders’

evaluations, it is important to demonstrate that high performance evaluations did not stem simply from a positive halo based on communication style or on how much each participant talked. Therefore, we coded some of the most commonly measured behaviors in dyadic interactions (i.e., speaking time, gazing, smiling).

We assessed how much each participant spoke during the 8-min interaction by summing up the duration of utterances of at least one word by stopwatch (in seconds). Reliability (2 coders) of speaking time was .96 (owners,M=161.94,SD=54.76; assistants,M=142.81,SD=54.01). How much each participant looked at the other was also measured by stopwatch (in seconds). Reliability (2 coders) of gazing was .82 (owners, M=23.95, SD=25.53; assistants, M=26.06, SD=32.75). Finally, we assessed how often each participant smiled during the 8-min interaction. Reliability (2 coders) of frequency of smiling was .92 (owners,M=12.99,SD=6.86; assistants,M=13.07,SD=6.48).

Results

To test how assigned power role and power preference affected dyadic performance motivation, we calculated a 2 (Power Preference of Owner:

owner vs. assistant)¥2 (Power Preference of Assistant: owner vs. assistant)

¥2 (Dyad Gender) ANOVA, with the performance motivation of the dyad as the dependent variable. The results show a significant power preference of owner main effect,F(1, 60)=6.61,p=.013 (effect size,r=.32), indicating that when owners had a preference for the owner role, the performance motivation of the dyad was higher than when owners had a preference for the assistant role. None of the other main or interaction effects was significant (all Fs<3.21, all ps>.08). The means are displayed in Table 1.

The same ANOVA calculated separately for the owners’ and assistants’

performance motivation confirmed this result and showed that for owners (and marginally so for assistants), performance motivation was higher when the owner wanted to be the owner, as compared to when the owner wanted to be the assistant: owners, F(1, 60)=7.34, p=.009; and assistants, F(1, 60)=3.80,p=.056. We also calculated the same ANOVA with the addition of a repeated-measures factor for role assignment (owner vs. assistant) to see whether the owner’s or assistant’s preference affected the performance motivation of the owner and the assistant differentially. This analysis again showed the aforementioned power preference of owner main effect, F(1, 60)=6.61,p=.013 (effect size,r=.32).

(10)

The interaction of owner’s preference and role assignment was not significant,F(1, 60)=1.17,p=.28, from which we can conclude that the owner’s preference affected the performance motivation of both owner and assistant in the same way. The interaction of assistant’s preference and role assignment, however, was significant, F(1, 60)=4.24, p=.044, showing that the assistant’s preference had a larger impact on his or her own performance motivation (M=3.01 when wanting to be the assistant; M=3.38 when wanting to be the owner) than on the owner’s performance motivation (M=3.23 when the assistant wanted to be the assistant;M=3.48 when the assistant wanted to be the owner).

To test whether speaking time, gazing, and smiling of both the owner and the assistant mediated the effects of power preference on dyadic performance motivation, we calculated a 2 (Power Preference of Owner: owner vs. assistant)¥2 (Power Preference of Assistant: owner vs. assistant) ANOVA, with performance motivation of the dyad as the dependent variable (leaving out gender of the dyad because it did not affect the results) while controlling for the three behaviors that were significantly related to rated performance motivation (speaking time of the owner, r=.24, p=.049; speaking time of the assistant,r=.55,p<.0001; gazing of the owner,r=.32,p=.009) by entering them as covariates in the analysis. The results show again that dyads in which the owner wanted to be the owner had better performance motivation than did dyads in which the owner wanted to be the assistant, F(1, 61)=3.43, p=.069 (effect size,r=.23). This effect was a little less pronounced, but still comparable to the one obtained from the ANOVA in which the other behaviors of the owner and the assistant were not controlled for (r=.32). We can conclude from this analysis that although talking and gazing were related to higher performance motivation ratings, it is not more talking or gazing that explains the fact that the owner’s preference predicted the dyad’s performance motivation.

Table 1

Means for Dyadic Performance Motivation

Owner’s preference

Assistant’s preference Assistant Owner

Assistant 2.84 3.55

Owner 3.31 3.73

Irrespective of assistant preference 3.08 3.64

(11)

Discussion

The goal of the present study was to investigate how power motivation of the superior and of the subordinate affect performance motivation in a dyad with a hierarchical structure. We predicted that dyads with a high power motivation leader (i.e., a leader who wanted to be leader) would have more pronounced performance motivation than would dyads with a low power motivation leader (i.e., a leader who wanted to be subordinate).

The results confirm this prediction in that the performance motivation (of each dyad member individually and of the dyad as a whole) depended on the boss’s power motivation. Bosses (i.e., assigned owners) who wanted to be in the high power position, as compared to bosses who wanted to be in the low power position, were responsible for more performance motivation both in themselves and in their assigned assistants. In contrast, power motivation of the subordinates did not affect dyadic performance motivation.

Power motivation of the subordinate affected the effort put into the task (i.e., performance motivation) of the subordinate. Subordinates with high power motivation invested relatively more effort than did subordinates with low power motivation, whereas the subordinate’s power motivation affected the owner’s effort to a significantly less pronounced degree. This finding goes against the prediction derived from leader–member exchange theory stating that subordinates with high power motivation might sabotage the relationship with the superior and thus put less effort into the task. The contrary was true; they put in relatively more effort.

The formation of hierarchies is ubiquitous (Bales, 1950; Berger, Con- ner, & Fisek, 1974; Tiedens & Fragale, 2003). Hierarchies are considered useful because they render a group more productive. The core idea is that established hierarchies prevent group members from having to negotiate the power relations within the group, which sets free the energy and cognitive capacity of each group member to focus on the task at hand, therefore enhancing performance.

Based on this reasoning, one explanation of the effect of power motivation of the leader on performance motivation might be that in dyads with owners who were motivated to lead, the hierarchy between the high and low power position might have been more salient. This could have made it easier for both participants to focus on and put effort into the task. The leader who wanted to be in the leadership position took on that role more convincingly than did the leader who was averse to the high power role (indicated by not wanting that role initially); therefore, less time was invested in defining the roles, and more resources could be used to do a better job on the task by both interaction partners. This corresponds to what is seen in an

(12)

expectation states theory approach as the superior being legitimate in his or her position.

In our study, the assistant’s preference was unrelated to dyadic performance motivation. This finding is reminiscent of work showing that powerful leaders affect subordinate performance to a far greater extent than powerful subordinates affect leader performance (e.g., Herold, 1977). Maybe because the superior usually has more responsibility for the dyad’s good task performance than does the subordinate, the subordinate’s characteristics play a less pronounced role.

Nevertheless, the subordinate’s power motivation was related to his or her individual performance scores. When subordinates were motivated to be superiors, their investment in the task was more pronounced than when their subordinate positions corresponded to what they wished for themselves: low power. Thus, although there was no gain in overall performance motivation for the dyad when subordinates aspired for the high power position, the subordinate’s motivation to do a good job still increased. Maybe subordinates who want to be in a high power position put extra effort into the task at hand to try to move up in the hierarchy to the desired high power position, at least in the eyes of the social interaction partner or in the eyes of the experimenter.

In real life, this might indeed be an effective strategy to attain higher ranks.

Superiors might not profit as much from subordinates’ extra efforts for their own performance motivation because superiors might be irritated by the fact that the subordinates do not adhere to the low power role.

We tested whether behaviors such as speaking time, gazing, and smiling of the owner and of the assistant would mediate the relation between the superior’s power motivation and dyadic performance motivation. Whereas speaking time of the owner and of the assistant, as well as gazing of the owner were positively related to dyadic performance motivation, these variables did not explain why dyads with high power motivation superiors invested more effort into the task (as reflected by receiving higher ratings of performance motivation) than did dyads with low power motivation superiors. This is in line with Schmid Mast and Hall (2003), who found that power motivation of the superior does not affect the superior’s amount of talk.

The present study has a number of limitations. Using an experimental design enabled us to investigate the sole or joint influence of power position and power motivation on performance motivation. Power position and power motivation are usually confounded in the real world, in that it is much more likely to find a person in a leadership position who wants to be in this high power position than a leader who does not want his or her leadership position.

One could argue that creating leaders who do not want to be leaders has low ecological validity. We think not, because there exists an analogue of a person

(13)

in a high power position who is not motivated to lead (i.e., owners who wanted to be assistants) in the real workplace: people who accept promotions to high power jobs not necessarily because they want to lead but because the new job offers other attractive features, such as a better salary or a more interesting work environment (e.g., going abroad). Moreover, promotions are often based on specific task competencies (e.g., technical skills), rather than on leadership qualities (Hogan, Curphy, & Hogan, 1994).This is not to say that competence should not enter the equation, but perhaps not as the sole factor.

Also, there are differences in personality dominance among leaders, and our design paralleled this with leaders who wanted or did not want to be in the leadership position. Note that the preference for a high or low power position (i.e., power motivation) was related to personality dominance.

Our study did not assess an objective performance measure. Rather, it measured performance motivation, which is defined as the effort put into the task and the quality of the task contributions. However, note that putting a focus on performance motivation in our study is reminiscent of the documented importance of the relationship between superiors and their subordinates concerning the motivation, satisfaction, and productivity of subordinates (Pincus, 1986).

Power motivation was operationalized as a preference for a superior or subordinate role and was related to personality dominance. However, it might be the case that certain aspects of power motivation—for instance, belief in one’s task abilities (i.e., felt task competence) or the feeling of greater responsibility—might be driving the effect. Moreover, we do not know whether indicating a preference for the subordinate role is an expression of wanting to be subordinate or an expression of avoiding the leadership role;

these two aspects are not necessarily identical.

The experimental nature of our study with short-term assignment of superior and subordinate roles, of course, does not take into account other factors (e.g., quality of the relationship between the boss and his or her subordinates, age gap often found between superiors and their subordinates) that are important in an actual hierarchical relationship in the workplace.

However, randomly assigning participants to the high versus low power roles and thus being able to draw causal inferences about the effect of power positions on outcome variables is a definite advantage of the present study.

Our findings have practical implications in that in real life, people some- times take promotions reluctantly or for reasons other than being the one in charge. We showed that the superior’s willingness (or lack thereof) to take on the responsibility of being a leader plays an important role for subsequent performance: Only the leaders who want to be in a higher power position lead more effectively. Our results also suggest that a subordinate who is motivated to achieve a higher power position will be particularly productive.

(14)

To conclude, it seems that dyadic performance motivation is best if the power hierarchy is clearly established, and thus is legitimated. In particular, it is important for the leader to embrace the power that goes with the leadership position. Though high power motivation is often seen in a rather negative light (i.e., power corrupts; Kipnis, 1972), our results show that wanting to be a boss is important for good leadership.

References

Bales, R. F. (1950). Interaction process analysis: A method for the study of small groups. Cambridge, MA: Addison-Wesley.

Barrick, M. R., & Mount, M. K. (1991). The Big Five personality dimensions and job performance: A meta-analysis.Personnel Psychology,44, 1–26.

Bass, B. M. (1985). Leadership and performance beyond expectation. New York: Free Press.

Berger, J., Conner, T. L., & Fisek, M. H. (1974).Expectation states theory: A theoretical research program. Cambridge, MA: Winthrop.

Berger, J., Ridgeway, C., Fisek, M. H., & Norman, R. Z. (1998). The legitimation and delegitimation of power and prestige orders. American Sociological Review,63, 379–405.

Bono, J. E., & Judge, T. A. (2003a). Core self-evaluations: A review of the trait and its role in job satisfaction and job performance. European Journal of Personality,17, 5–18.

Bono, J. E., & Judge, T. A. (2003b). Self-concordance at work: Toward understanding the motivational effects of transformational leaders.

Academy of Management Journal,46, 554–571.

Burns, J. M. (1978).Leadership. New York: Harper & Row.

Chan, K. Y., & Drasgow, F. (2001). Toward a theory of individual differences and leadership: Understanding the motivation to lead.Journal of Applied Psychology,86, 481–498.

Charbonneau, D., Barling, J., & Kelloway, E. K. (2001). Transformational leadership and sports performance: The mediating role of intrinsic motivation.Journal of Applied Social Psychology,31, 1521–1534.

Chidester, T. R., Helmreich, R. L., Gregorich, S. E., & Geis, C. E. (1991).

Pilot personality and crew coordination: Implications for training and selection.International Journal of Aviation Psychology,1, 25–44.

Gerstner, C. R., & Day, D. V. (1997). Meta-analytic review of leader–

member exchange theory: Correlates and construct issues. Journal of Applied Psychology,82, 827–844.

Graen, G. B., Novak, J. A., & Sommerkamp, P. (1982). The effects of leader–member exchange and job design on productivity and satisfaction:

(15)

Testing a dual attachment model. Organizational Behavior and Human Performance,30, 109–131.

Gough, H. G. (1975).Manual for the California Psychological Inventory. Palo Alto, CA: Consulting Psychologists Press.

Herold, D. M. (1977). Two-way influence processes in leader–follower dyads.

Academy of Management Journal,20, 224–237.

Hogan, R., Curphy, G. J., & Hogan, J. (1994). What we know about leadership: Effectiveness and personality.American Psychologist,49, 493–504.

Hoyt, C. L., & Blascovich, J. (2003). Transformational and transactional leadership in virtual and physical environments. Small Group Research, 34, 678–715.

Judge, T. A., & Bono, J. E. (2001). Relationship of core self-evaluations traits—self-esteem, generalized self-efficacy, locus of control, and emotional stability—with job satisfaction and job performance: A meta- analysis.Journal of Applied Psychology,86, 80–92.

Judge, T. A., Bono, J. E., Ilies, R., & Gerhardt, M. W. (2002). Personality and leadership: A qualitative and quantitative review.Journal of Applied Psychology,87, 765–780.

Keller, R. T. (2006). Transformational leadership, initiating structure, and substitutes for leadership: A longitudinal study of research and develop- ment project team performance.Journal of Applied Psychology,91, 202–

210.

Kipnis, D. (1972). Does power corrupt? Journal of Personality and Social Psychology,24, 33–41.

Klimoski, R. J., & Hayes, N. J. (1980). Leader behavior and subordinate motivation.Personnel Psychology,33, 543–555.

Lord, R. G., & Hall, R. J. (1992). Contemporary views of leadership and individual differences.Leadership Quarterly,3, 137–157.

Mann, R. D. (1959). A review of the relationships between personality and performance in small groups.Psychological Bulletin,56, 241–270.

McColl-Kennedy, J. R., & Anderson, R. D. (2002). Impact of leadership style and emotions on subordinate performance. Leadership Quarterly, 13, 545–559.

Pincus, J. D. (1986). Communication satisfaction, job satisfaction, and job performance.Human Communication Research,12, 395–419.

Ridgeway, C. L., & Berger, J. (1986). Expectations, legitimation, and dominance behavior in task groups. American Sociological Review, 51, 603–

617.

Schmid Mast, M., & Hall, J. A. (2003). Anybody can be a boss, but only certain people make good subordinates: Behavioral impacts of striving for dominance and dominance aversion.Journal of Personality,71, 871–

891.

(16)

Schutz, W. C. (1958). FIRO: A three-dimensional theory of interpersonal behavior. New York: Rinehart.

Shamir, B., House, R. J., & Arthur, M. B. (1993). The motivational effects of charismatic leadership: A self-concept based theory. Organization Science,4, 577–594.

Tiedens, L. Z., & Fragale, A. R. (2003). Power moves: Complementary in submissive and dominant nonverbal behavior.Journal of Personality and Social Psychology,84, 558–568.